TY - JOUR
T1 - Prediction of ${\rm PM}_{2.5}$ Concentration in Beijing Based on Bayesian Hierarchical Autoregressive Spatio-Temporal Model
AU - Wang , Jing
AU - Cao , Chunzheng
JO - Journal of Information and Computing Science
VL - 2
SP - 117
EP - 128
PY - 2023
DA - 2023/12
SN - 18
DO - http://doi.org/10.4208/JICS-2023-008
UR - https://global-sci.org/intro/article_detail/jics/23724.html
KW - Bayesian method, Hierarchical model, Autoregressive, Spatio-temporal model, ${\rm PM}_{2.5}$ prediction, Markov Chain Monte Carlo (MCMC).
AB - <p style="text-align: justify;">Here, a hierarchical autoregressive spatio-temporal model under the Bayesian
framework is proposed to address the simultaneous multi-site <span style="text-align: justify; text-wrap-mode: wrap;">${\rm PM}_{2.5}$</span> prediction. The true
daily average concentration of ${\rm PM}_{2.5}$ is regarded as a potential spatio-temporal process,
then the temporal correlation is described by the first-order autoregressive process and
the spatial correlation is captured based on the Matérn process, which greatly improves
the efficiency in dimension reduction and synchronous prediction. In addition,
meteorological factors such as daily maximum temperature, relative humidity and wind
speed are used as explanatory variables to improve the prediction accuracy. The
combination of Bayesian method and MCMC can realize parameter estimation and
prediction process due to the model&#39;s hierarchical structure. The empirical analysis of
daily <span style="text-align: justify; text-wrap-mode: wrap;">${\rm PM}_{2.5}$</span> concentration in Beijing shows that the proposed model has good
interpolation or prediction performance in both spatial and temporal dimensions.</p>